Biologizing the stack

I’m experimenting with writing some shorter, less polished posts. These come with the disclaimer that I’m optimizing for getting thoughts out there while they’re still crystallizing rather than achieving maximum rigor. For this post in particular, I’m not arguing that this direction is definitely worth pursuing so much as exploring a frame/idea to see where it gets me.

A while ago, a friend asked me why I like the idea of working with cells better than cell-free systems and I gave some hand-wavy answer about cells being cool. Some of Elliot Hershberg’s not so recent and more recent posts got me thinking about this question again and I think I’ve figured out a more satisfying answer that also captures the theme in some of the ideas in both my and Arye’s old post.

Expanding upon the idea of “biologization of industry” from Drew Endy, Elliot lays out an exciting vision of local production of therapeutics and materials using biology. Elliot describes an antibody printer which can print antibodies for any use-case and a “biomaker” capable of constructing biological organisms from scratch. First off, I love Elliot’s post and the philosophy it embraces. Biotech needs much more vision-setting like Elliot’s done here, especially to compete with those pesky AGI folks.

Rereading the post though, I noticed that a lot of the pictures of both the hypothetical artifacts and existing systems Elliot describes are essentially plastic boxes of electronics with a little biology in a carefully controlled container or chip. To be fair, add in some plates and bioreactors and this mirrors how much of synthetic biology happens today. Biology in the lab seeks to sterilize, isolate, and control individual steps as much as possible.

This got me thinking what it would look like to move towards a “fully biological stack”. Defined extensionally, in the fully biological stack:

• Rather than boxes, we have designed plants that produces therapeutic antibodies in their fruit or seeds
• Rather than an alien dreadnought, the ultimate form of a lab is a programmable alien organism, with things like helicases replacing thermocyclers representing small steps on the path
• Rather than robot surgeons, anatomical compilers
• Cell therapies become organ therapies become symbiote therapies
• Electronic continuous monitoring devices get replaced by molecular records
• Cell-based meat moves from vats to custom organoids Summarizing, the fully biological stack relies on biological components and principles in as many steps as possible while embracing self-organization and and interconnectedness.

Obviously all this sounds very biopunk and fun, but it also probably sounds like a pipe dream or seems like a false dichotomy. Especially in light of the enchippening thesis of recent technological progress, this is not only a grand vision but a contrarian one. Rather than arguing for more of the thing that’s got us to where we are today – using our knowledge of physics to introspect and control biology – I’m arguing for trying to get biology to do more stuff itself. (Side note: Although in reality, it’s a false dichotomy.) Why might this direction still be worth pushing given it involves diverging from powerful rising tides of progress?

So far, the best reason I’ve come up with is that this direction may be necessary for achieving the more ambitious goals of synthetic biology that initially attracted so many people to the movement. As Arye Lipman discusses in Set biotech free, our current paradigm relies on maintaining sterility, environmental control, and avoiding system heterogeneity where possible, crutches that evolved biology eschews entirely. Biological systems in the wild survive and thrive in much messier, less controlled environments, and they evolve and change to adapt to their environments. If we want to grow houses, grow meat at scale, make dragons, or biologize industry, we need to figure out how to build biological systems that thrive in messy environments and make complexity work for us rather than against us. At least the journey towards biologizing the stack seems likely to help us learn how to do this.

So, bringing things all the way back to the original question, the reason I’m drawn to cells is because the only fully biological stack we know seems to find higher level organization with membranes and morphology useful. If we want to achieve the same power, adaptability, environmental robustness, and complexity that nature has, we probably need to embrace rather than discard the tools it’s found for doing so.